分布式滞后神经加性模型
Distributed Lag Neural Additive Models
浏览论文内容
中文总结 AI 辅助
提出分布式滞后神经加性模型（DLNAMs），用神经组件替代样条交叉基学习暴露-滞后非线性效应，在模拟中优于DLNM比较模型，具有更低偏差和更好校准。
中文摘要 AI 辅助
我们提出了分布式滞后神经加性模型（DLNAMs），这是分布式滞后非线性模型（DLNMs）的神经加性对应物，用于学习在滞后上分布的非线性效应。DLNAMs 用神经组件替代预先指定的样条交叉基，这些神经组件学习暴露-滞后响应曲面，避免了基函数族、维度和节点位置的选择，同时保留了加性可解释性和熟悉的分布式滞后汇总。以指数为中心的输入层、平滑激活函数和学习的子网络混合产生平滑、局部自适应的表示；逐点不确定性结合了条件最后一层拉普拉斯近似与成员间集成变异。在模拟中，DLNAMs 在恢复已知响应函数方面通常优于 DLNM 比较模型（包括惩罚和树变体），具有更低的偏差、更强的边界恢复和更好的校准累积区间；对于更具挑战性的函数，收益最大。该架构在样本量、结果族、滞后范围以及联合拟合的多暴露设置中表现一致，随着暴露的增加保持恢复性能；拟合特定的变化主要限于优化，应用恢复了既定的经验模式。
英文摘要
We introduce Distributed Lag Neural Additive Models (DLNAMs), neural-additive analogues of Distributed Lag Non-linear Models (DLNMs) for learning nonlinear effects distributed over lags. DLNAMs replace a prespecified spline cross-basis with neural components that learn exposure--lag response surfaces, avoiding choices of basis family, dimension, and knot placement while preserving additive interpretability and familiar distributed-lag summaries. Exp-centered input layers, smooth activations, and learned subnetwork mixtures produce smooth, locally adaptive representations; pointwise uncertainty combines a conditional last-layer Laplace approximation with between-member ensemble variation. In simulations, DLNAMs generally outperformed DLNM comparators, including penalized and treed variants, in recovering known response functions, with lower bias, stronger boundary recovery, and better-calibrated cumulative intervals; gains were largest for more demanding functions. The architecture performed consistently across sample sizes, outcome families, lag horizons, and jointly fitted multi-exposure settings, retaining recovery performance as exposures were added; fit-specific changes were largely confined to optimization, and applications recovered established empirical patterns.
发表机构
- KTH Royal Institute of Technology（KTH皇家理工学院）
- Karolinska Institutet（卡罗林斯卡学院）
- Uppsala University（乌普萨拉大学）
- University of Cambridge（剑桥大学）
机构由 AI 辅助整理，请以论文原文为准。
